Data-Driven Explainable Machine Learning Approaches for Predicting Hydrogen Adsorption in Porous Crystalline Materials DOI
Hung Vo Thanh, Zhenxue Dai, Mohammad Rahimi

et al.

Journal of Alloys and Compounds, Journal Year: 2025, Volume and Issue: unknown, P. 180709 - 180709

Published: May 1, 2025

Language: Английский

Artificial intelligence-based prediction of hydrogen adsorption in various kerogen types: Implications for underground hydrogen storage and cleaner production DOI
Hung Vo Thanh, Zhenxue Dai,

Zhengyang Du

et al.

International Journal of Hydrogen Energy, Journal Year: 2024, Volume and Issue: 57, P. 1000 - 1009

Published: Jan. 13, 2024

Language: Английский

Citations

28
Catalyzing net-zero carbon strategies: Enhancing CO2 flux Prediction from underground coal fires using optimized machine learning models DOI

Hemeng Zhang,

Pengcheng Wang,

Mohammad Rahimi

et al.

Journal of Cleaner Production, Journal Year: 2024, Volume and Issue: 441, P. 141043 - 141043

Published: Jan. 31, 2024

Language: Английский

Citations

24
Advancing hydrogen storage predictions in metal-organic frameworks: A comparative study of LightGBM and random forest models with data enhancement DOI Creative Commons
Masoud Seyyedattar, Sohrab Zendehboudi, Ali Ghamartale

et al.

International Journal of Hydrogen Energy, Journal Year: 2024, Volume and Issue: 69, P. 158 - 172

Published: May 6, 2024

The escalating consumption of fossil fuels has given rise to a substantial upsurge in greenhouse gas concentrations and global temperatures, which, turn, triggered severe climate-related consequences. critical imperative reduce CO2 emissions combat warming spurred extensive investigations into clean energy alternatives, with hydrogen emerging as compelling zero-emission source. As pivotal component strategies, requires designing compact, lightweight, efficient storage systems. This study focuses on the development evaluation machine learning models for predicting efficiency Metal-Organic Frameworks (MOFs) storage, key aspect advancing technologies. MOFs, class nanoporous materials, show remarkable potential due their high surface area porosity. However, selecting most suitable MOF this application from vast array possible structures is daunting task. In context, algorithms offer an alternative suitability by considering structural chemical properties. We used ensemble methods, specifically Light Gradient Boosting Machine (LightGBM) Random Forest (RF), predict uptake MOFs based dataset 219 experimentally tested samples. Two modeling scenarios were considered: one using entire dataset, other involving strategic data pre-processing, including outlier removal feature engineering. results demonstrate that measures taken refine significantly enhance predictive performance developed models, reducing prediction errors improving overall goodness fit. Specifically, Mean Absolute Error (MAE) values both LightGBM random forest reduced 0.48 0.94, respectively, 0.16 coefficients determination (R2) increased substantially 0.84 0.72 0.95, cases. Moreover, importance analysis unveiled pressure-related features make significant contributions formation tree ensembles during model training process. A parametric sensitivity was conducted revealing H2 sensitive changes adsorption enthalpy, followed temperature, while showing lower variations pressure, consistent established literature. These underscore role enhancement methods refining can be instrumental accelerating optimization materials applications.

Language: Английский

Citations

19
Predicting Hydrogen Fuel Cell Capacity using Supervised Learning Models DOI

Vinay Nagarad Dasavandi Krishnamurthy,

Sheshang Degadwala, Dhairya Vyas

et al.

2022 International Conference on Inventive Computation Technologies (ICICT), Journal Year: 2024, Volume and Issue: unknown

Published: April 24, 2024

Hydrogen fuel cells have emerged as a promising solution for clean energy, but their effectiveness and reliability depend on the precise prediction of capacity. This research study investigates into application various supervised learning models to forecast hydrogen cell The findings uncover distinctive strengths limitations each regression model in context capacity prediction. Linear Regression stands out its simplicity transparency, offering an easy-to-understand approach. On other hand, Random Forest Decision Tree demonstrate knack handling non-linear relationships within data. KNN excels capturing localized patterns, while Gradient Boosting utilizes ensemble achieve heightened accuracy. SVR exhibits adaptability through kernel functions, Logistic proves effective binary classification tasks. Meanwhile, Polynomial effectively captures potential non-linearity present provides guidance choosing best certain scenarios by evaluating performance across range assessment measures, such mean squared error, absolute R-squared values.

Language: Английский

Citations

19
Predicting uniaxial compressive strength from drilling variables aided by hybrid machine learning DOI
Shadfar Davoodi, Mohammad Mehrad, David A. Wood

et al.

International Journal of Rock Mechanics and Mining Sciences, Journal Year: 2023, Volume and Issue: 170, P. 105546 - 105546

Published: July 17, 2023

Language: Английский

Citations

32
Machine learning in absorption-based post-combustion carbon capture systems: A state-of-the-art review DOI Creative Commons

Milad Hosseinpour,

Mohammad Javad Shojaei, Mohsen Salimi

et al.

Fuel, Journal Year: 2023, Volume and Issue: 353, P. 129265 - 129265

Published: July 25, 2023

Language: Английский

Citations

31
Machine learning insights to CO2-EOR and storage simulations through a five-spot pattern – a theoretical study DOI
Shadfar Davoodi, Hung Vo Thanh, David A. Wood

et al.

Expert Systems with Applications, Journal Year: 2024, Volume and Issue: 250, P. 123944 - 123944

Published: April 13, 2024

Language: Английский

Citations

11
Enhanced hydrogen storage efficiency with sorbents and machine learning: a review DOI Creative Commons
Ahmed I. Osman, Walaa Abd‐Elaziem, Mahmoud Nasr

et al.

Environmental Chemistry Letters, Journal Year: 2024, Volume and Issue: 22(4), P. 1703 - 1740

Published: May 16, 2024

Abstract Hydrogen is viewed as the future carbon–neutral fuel, yet hydrogen storage a key issue for developing economy because current techniques are expensive and potentially unsafe due to pressures reaching up 700 bar. As consequence, research has recently designed advanced sorbents, such metal–organic frameworks, covalent organic porous carbon-based adsorbents, zeolite, composites, safer storage. Here, we review with focus on sources production, machine learning. Carbon-based sorbents include graphene, fullerene, carbon nanotubes activated carbon. We observed that capacities reach 10 wt.% 6 3–5 adsorbents. High-entropy alloys composites exhibit improved stability uptake. Machine learning allowed predicting efficient materials.

Language: Английский

Citations

11
Recent advances in sustainable and efficient hydrogen storage nanomaterials DOI
Nour F. Attia, Sally E.A. Elashery,

Mohamed A. Nour

et al.

Journal of Energy Storage, Journal Year: 2024, Volume and Issue: 100, P. 113519 - 113519

Published: Aug. 29, 2024

Language: Английский

Citations

9
Low-Carbon Advancement through Cleaner Production: A Machine Learning Approach for Enhanced Hydrogen Storage Predictions in Coal Seams DOI
Yongjun Wang, Hung Vo Thanh,

Hemeng Zhang

et al.

Renewable Energy, Journal Year: 2025, Volume and Issue: unknown, P. 122342 - 122342

Published: Jan. 1, 2025

Language: Английский

Citations

1